Bill, thanks for the video. Very nice musical accompaniment!
I have reflected a bit on your theory of the swinging/centrifugal (actually, centripetal, but who's counting?) effect in OTT shooting and its contribution to increased variability. Without a doubt, the effect is there. However, would it be there before the ammo has left the pouch? I'm not so sure (perhaps fellow Canuck Charles will chime in on this as I remember him commenting somewhere on when the ammo leaves the pouch). The swinging effect is caused by the attachment point(s) pulling back on the bands. This occurs once the bands have passed through the forks and have extended in the other direction. For TTF, the attachment points are aligned with the target line, hence the pull is straight back. But in OTT shooting, the attachments points are offset from the target line, resulting in a small component of force to the left (for right-handed pullers), which causes the bands to swing left on their way back, resulting in potential hand slap. But again, I'm not sure this should affect accuracy as I imagine the ammo will have left the pouch long before the swinging component of force kicks in.
Consequently, I'll throw another idea out there. I wonder if the increased inconsistency in OTT shooting can be caused by the band attachment method. Looking down the top band, we have a very skinny sight line near the forks. But near the pouch, the bands on my shooter are folded in half in order to attach to the pouch. If we cut the folded parts of the bands in half with a vertical line (plane), what is on the left side of the plane is not the same as what is on the right, i.e., it's not symmetrical. I'm wondering if this assymetry might cause small left-right deviations from the target line.
In the end, this is probably all a moot point. Most OTT shooters will likely stick with OTT. But I'm a bit of a perfectionist. If there is a potential for better accuracy with TTF, then I'm going to try hard to learn TTF.